Tracers labeled with short-lived positron emitting radionuclides (e.g. 11C, t1/2=20.3 min) are frequently used in various non-invasive in vivo studies in combination with positron emission tomography (PET). Because of the radioactivity, the short half-lives and the submicromolar amounts of the labeled substances, extraordinary synthetic procedures are required for the production of these tracers. An important part of the elaboration of these procedures is development and handling of new 11C-labelled precursors. This is important not only for labeling new types of compounds, but also for increasing the possibility of labeling a given compound in different positions.
During the last two decades carbonylation chemistry using carbon monoxide has developed significantly. The recent development of methods such as palladium-catalyzed carbonylative coupling reactions has provided a mild and efficient tool for the transformation of carbon monoxide into different carbonyl compounds.
Carbonylation reactions using [11C]carbon monoxide has a primary value for PET-tracer synthesis since biologically active substances often contain carbonyl groups or functionalities that can be derived from a carbonyl group. The syntheses are tolerant to most functional groups, which means that complex building blocks can be assembled in the carbonylation step to yield the target compound. This is particularly valuable in PET-tracer synthesis where the unlabelled substrates should be combined with the labeled precursor as late as possible in the reaction sequence, in order to decrease synthesis-time and thus optimize the uncorrected radiochemical yield.
When compounds are labeled with 11C, it is usually important to maximize specific radioactivity. In order to achieve this, the isotopic dilution and the synthesis time must be minimized. Isotopic dilution from atmospheric carbon dioxide may be substantial when [11C]carbon dioxide is used in a labeling reaction. Due to the low reactivity and atmospheric concentration of carbon monoxide (0.1 ppm vs. 3.4×104 ppm for CO2), this problem is reduced with reactions using [11C]carbon monoxide.
The synthesis of [11C]carbon monoxide from [11C]carbon dioxide using a heated column containing reducing agents such as zinc, charcoal or molybdenum has been described previously in several publications. Although [11C]carbon monoxide was one of the first 11C-labelled compounds to be applied in tracer experiments in human, it has until recently not found any practical use in the production of PET-tracers. One reason for this is the low solubility and relative slow reaction rate of [11C]carbon monoxide which causes low trapping efficiency in reaction media. The general procedure using precursors such as [11C]methyl iodide, [11C]hydrogen cyanide or [11C]carbon dioxide is to transfer the radioactivity in a gas-phase, and trap the radioactivity by leading the gas stream through a reaction medium. Until recently this has been the only accessible procedure to handle [11C]carbon monoxide in labeling synthesis. With this approach, the main part of the labeling syntheses with [11C]carbon monoxide can be expected to give a very low yield or fail completely.
There are only a few examples of practically valuable 11C-labelling syntheses using high pressure techniques (>300 bar). In principal, high pressures can be utilized for increasing reaction rates and minimizing the amounts of reagents. One problem with this approach is how to confine the labeled precursor in a small high-pressure reactor. Another problem is the construction of the reactor. If a common column type of reactor is used (i.e. a cylinder with tubing attached to each end), the gas-phase will actually become efficiently excluded from the liquid phase at pressurization. The reason is that the gas-phase, in contracted form, will escape into the attached tubing and away from the bulk amount of the liquid reagent.
The cold-trap technique is widely used in the handling of 11C-labelled precursors, particularly in the case of [11C]carbon dioxide. The procedure has, however, only been performed in one single step and the labeled compound was always released in a continuous gas-stream simultaneous with the heating of the cold-trap. Furthermore, the volume of the material used to trap the labeled compound has been relative large in relation to the system to which the labeled compound has been transferred. Thus, the option of using this technique for radical concentration of the labeled compound and miniaturization of synthesis systems has not been explored. This is especially noteworthy in view of the fact that the amount of a 11C-labelled compound usually is in the range 20-60 nmol.
Recent technical development for the production and use of [11C] carbon monoxide has made this compound useful in labeling synthesis. WO 02/102711 describes a system and a method for the production and use of a carbon-isotope monoxide enriched gas-mixture from an initial carbon-isotope dioxide gas mixture. [11C] carbon monoxide may be obtained in high radiochemical yield from cyclotron produced [11C] carbon dioxide and can be used to yield target compounds with high specific radioactivity. This reactor overcomes the difficulties listed above and is useful in synthesis of 11C-labelled compounds using [11C] carbon monoxide in palladium or selenium mediated reaction. With such method, a broad array of carbonyl compounds can be labeled (Kilhlberg, T.; Langstrom, B. J., Org. Chem. 1999, 9201-9205). The use of transition metal mediated reactions is, however, restricted by problems related to the competing β-hydride elimination reaction, which excludes or at least severely restricts utilization of organic electrophiles having hydrogen in β-position. Thus, a limitation of the transition metal mediated reactions is that most alkyl halides could not be used as substrates due to the 13-hydride elimination reaction. One way to circumvent this problem is to use free-radical chemistry based on light irradiation of alkyl halides. We earlier succeeded in using free-radical chemistry for the carbonylation of alkyl iodides using amines to yield labeled amides. However, the attempt to yield labeled esters in an analogous way (using alcohols as a reactant instead of amines) is challenged by the low reactivity of alcohols in these reaction conditions (typically the yields of esters compared to those of amides are lower by 10 to 100 times). Therefore, there is a need for a method in order to use photo-induced free radical carbonylation with weakly reacting alcohols to circumvent the problem with β-hydride elimination to complement the palladium mediated reactions and provide target structures with high yield to further increase the utility of [11C] carbon monoxide in preparing useful PET tracers.
Discussion or citation of a reference herein shall not be construed as an admission that such reference is prior art to the present invention.